tools for working with chialisp language; compiler, repl, python and wasm bindings
Project description
clvm_tools_rs
Theory of operation of the modern compiler: ./HOW_CHIALISP_IS_COMPILED.md
This repo can be installed via cargo
cargo install clvm_tools_rs
or via pip
pip install clvm_tools_rs@git+https://github.com/Chia-Network/clvm_tools_rs.git@e17412032aa7d3b8b1d1f931893fb5802eee626a
Note: pip
installs a subset of the tools installed by cargo
, including brun
, run
, opc
and opd
.
The most current version of the language is in the nightly branch:
[nightly](https://github.com/Chia-Network/clvm_tools_rs/tree/nightly)
To install from a specific branch:
cargo install --no-default-features --git 'https://github.com/Chia-Network/clvm_tools_rs' --branch nightly
To install a git checkout into your current python environment (must be in some kind of venv or conda environment):
git clone https://github.com/Chia-Network/clvm_tools_rs
cd clvm_tools_rs
maturin develop
Install from PYPI:
pip install -i https://pypi.chia.net/nightlies/ clvm_tools_rs
Most people still compile chialisp via python. One way to set up compilation in that way is like this:
import json
from clvm_tools_rs import compile_clvm
def compile_module_with_symbols(include_paths,source):
path_obj = Path(source)
file_path = path_obj.parent
file_stem = path_obj.stem
target_file = file_path / (file_stem + ".clvm.hex")
sym_file = file_path / (file_stem + ".sym")
compile_result = compile_clvm(source, str(target_file.absolute()), include_paths, True)
symbols = compile_result['symbols']
if len(symbols) != 0:
with open(str(sym_file.absolute()),'w') as symfile:
symfile.write(json.dumps(symbols))
The command line tools provided:
- run -- Compiles CLVM code from chialisp
Most commonly, you'll compile chialisp like this:
./target/debug/run -O -i include_dir chialisp.clsp
'run' outputs the code resulting from compiling the program, or an error.
- repl -- Accepts chialisp forms and expressions and produces results
interactively.
Run like:
./target/debug/repl
Example session:
>>> (defmacro assert items
(if (r items)
(list if (f items) (c assert (r items)) (q . (x)))
(f items)
)
)
(q)
>>> (assert 1 1 "hello")
(q . hello)
>>> (assert 1 0 "bye")
failed: CompileErr(Srcloc { file: "*macros*", line: 2, col: 26, until: Some(Until { line: 2, col: 82 }) }, "clvm raise in (8) (())")
>>>
- cldb -- Stepwise run chialisp programs with program readable yaml output.
./target/debug/cldb '(mod (X) (x X))' '(4)'
---
- Arguments: (() (4))
Operator: "4"
Operator-Location: "*command*(1):11"
Result-Location: "*command*(1):11"
Row: "0"
Value: (() 4)
- Env: "4"
Env-Args: ()
Operator: "2"
Operator-Location: "*command*(1):11"
Result-Location: "*command*(1):13"
Row: "1"
Value: "4"
- Arguments: (4)
Failure: clvm raise in (8 5) (() 4)
Failure-Location: "*command*(1):11"
Operator: "8"
Operator-Location: "*command*(1):13"
- brun -- Runs a "binary" program. Instead of serving as a chialisp
compiler, instead runs clvm programs.
As 'brun' from the python code:
$ ./target/debug/run '(mod (X) (defun fact (N X) (if (> 2 X) N (fact (* X N) (- X 1)))) (fact 1 X))'
(a (q 2 2 (c 2 (c (q . 1) (c 5 ())))) (c (q 2 (i (> (q . 2) 11) (q . 5) (q 2 2 (c 2 (c (* 11 5) (c (- 11 (q . 1)) ()))))) 1) 1))
$ ./target/debug/brun '(a (q 2 2 (c 2 (c (q . 1) (c 5 ())))) (c (q 2 (i (> (q . 2) 11) (q . 5) (q 2 2 (c 2 (c (* 11 5) (c (- 11 (q . 1)) ()))))) 1) 1))' '(5)'
120
- opc -- crush clvm s-expression form to hex.
As 'opc' from the python code.
opc '(a (q 2 2 (c 2 (c (q . 1) (c 5 ())))) (c (q 2 (i (> (q . 2) 11) (q . 5) (q 2 2 (c 2 (c (* 11 5) (c (- 11 (q . 1)) ()))))) 1) 1))'
ff02ffff01ff02ff02ffff04ff02ffff04ffff0101ffff04ff05ff8080808080ffff04ffff01ff02ffff03ffff15ffff0102ff0b80ffff0105ffff01ff02ff02ffff04ff02ffff04ffff12ff0bff0580ffff04ffff11ff0bffff010180ff808080808080ff0180ff018080
- opd -- disassemble hex to s-expression form.
As 'opd' from the python code.
opd 'ff02ffff01ff02ff02ffff04ff02ffff04ffff0101ffff04ff05ff8080808080ffff04ffff01ff02ffff03ffff15ffff0102ff0b80ffff0105ffff01ff02ff02ffff04ff02ffff04ffff12ff0bff0580ffff04ffff11ff0bffff010180ff808080808080ff0180ff018080'
(a (q 2 2 (c 2 (c (q . 1) (c 5 ())))) (c (q 2 (i (> (q . 2) 11) (q . 5) (q 2 2 (c 2 (c (* 11 5) (c (- 11 (q . 1)) ()))))) 1) 1))
History
This is a second-hand port of chia's clvm tools to rust via the work of ChiaMineJP porting to typescript. This would have been a lot harder to get to where it is without prior work mapping out the types of various semi-dynamic things (thanks, ChiaMineJP).
Some reasons for doing this are:
-
Chia switched the clvm implementation to rust: clvm_rs, and this code may both pick up speed and track clvm better being in the same language.
-
I wrote a new compiler with a simpler, less intricate structure that should be easier to improve and verify in the future in ocaml: ochialisp.
-
Also it's faster even in this unoptimized form.
All acceptance tests i've brought over so far work, and more are being added. As of now, I'm not aware of anything that shouldn't be authentic when running these command line tools from clvm_tools in their equivalents in this repository
-
opc
-
opd
-
run
-
brun
-
repl
argparse was ported to javascript and I believe I have faithfully reproduced it as it is used in cmds, so command line parsing should work similarly in all three versions.
The directory structure is expected to be:
src/classic <-- any ported code with heritage pointing back to
the original chia repo.
src/compiler <-- a newer compiler (ochialisp) with a simpler
structure. Select new style compilation by
including a `(include *standard-cl-21*)`
form in your toplevel `mod` form.
Mac M1
Use cargo build --no-default-features
due to differences in how mac m1 and
other platforms handle python extensions.
Use with chia-blockchain
# Activate your venv, then
$ maturin develop --release
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distributions
File details
Details for the file clvm_tools_rs-0.1.45.tar.gz
.
File metadata
- Download URL: clvm_tools_rs-0.1.45.tar.gz
- Upload date:
- Size: 549.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 73848e0504e42bbd9627497c7307147cce9a04678936b5aec5c5a6be4b372c84 |
|
MD5 | cb5053fa0242e14fe380a7580cc10269 |
|
BLAKE2b-256 | ba611369700ba72256b0ffaeb5b6407545b2b3fd3ea6a990687511828d8cc4b0 |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-win_amd64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-win_amd64.whl
- Upload date:
- Size: 1.4 MB
- Tags: CPython 3.8+, Windows x86-64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3153335cafd180a6a308fef0d3808777fb5f7e13edd43e9ee344afd212e05516 |
|
MD5 | df4cd0597fcf74a0a15bc6b1d1524370 |
|
BLAKE2b-256 | 057e5a0c10181cee6aa5e3538bb84ba11e285e1ba2af6a6520e29b1e559b1c34 |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-musllinux_1_1_x86_64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-musllinux_1_1_x86_64.whl
- Upload date:
- Size: 1.6 MB
- Tags: CPython 3.8+, musllinux: musl 1.1+ x86-64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 079b849e0c289b6c8e7e93c027d9ee8083714ad07c5d19a1c0f7bd88efa9615f |
|
MD5 | 51132f338387e8acb286dc104c100a2a |
|
BLAKE2b-256 | ad0aceadddaa13a971c54eee29a4d803dce4c30824671caf485ab52673028cbe |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-manylinux_2_28_x86_64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-manylinux_2_28_x86_64.whl
- Upload date:
- Size: 1.5 MB
- Tags: CPython 3.8+, manylinux: glibc 2.28+ x86-64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 792fb5be40d317e32b86e44f9a3bbbf4d96827b304877e678eaff0abca212b3e |
|
MD5 | d174110398042bc40e2276326ca3c9bf |
|
BLAKE2b-256 | 19865d5c34ac56f7d1241f41cf49e266c4c504115d4c71da8d8ff80b58ba73b4 |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-manylinux_2_28_aarch64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-manylinux_2_28_aarch64.whl
- Upload date:
- Size: 1.5 MB
- Tags: CPython 3.8+, manylinux: glibc 2.28+ ARM64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | fe0dcb7cad0edd1c9b1f7b0e3e6864cea1e13c3f8a6667de83c4721ff5626b4f |
|
MD5 | 2cda4976ae4d16e47784fda84635ea94 |
|
BLAKE2b-256 | ed63d5184653ec03f78ca831b4dfaa7c2f814b1709257253f15202345e6f2a30 |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-macosx_13_0_x86_64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-macosx_13_0_x86_64.whl
- Upload date:
- Size: 1.4 MB
- Tags: CPython 3.8+, macOS 13.0+ x86-64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | d1cd3240fdb9560c5cc860ac15196a00b70935827018deacf01651fcd9bfa6e5 |
|
MD5 | fc6f9821c921f92b4a350b9608b86a6d |
|
BLAKE2b-256 | a54e53e7f79f877b1de101829da203419f09bd02c418485918f6d1040e73549a |
File details
Details for the file clvm_tools_rs-0.1.45-cp38-abi3-macosx_11_0_arm64.whl
.
File metadata
- Download URL: clvm_tools_rs-0.1.45-cp38-abi3-macosx_11_0_arm64.whl
- Upload date:
- Size: 1.3 MB
- Tags: CPython 3.8+, macOS 11.0+ ARM64
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | fc5a60fa6bbbfb8875258266b82251c4cc9279e3d6435eaaada0b2c033816b1a |
|
MD5 | ce122a9a232180ea4ce9cc95dcd9bbd6 |
|
BLAKE2b-256 | 53abc0996508ae9cf855e18c20804fcbdabc518183b89a68db127039b1ac920d |